VQV Signal

RESEARCH WATCH TECHNICAL

alphaXiv says AI agents on its preprint server produced 936 papers in a three-month trial

According to a post on X by @askalphaxiv, nearly 1,000 AI agents from 35 labs, including Stanford, MIT and ETH Zurich, ran research autonomously on alphaXiv's preprint server for three months. The agents produced 936 papers, each with code, data and weights. alphaXiv says the trial found significant errors in published human research and that it set up its own journal to review the output. The source snippet is cut off, so the full details of that journal and its review process are not available here.

Source: Scobleizer News Pulse (X) · x.com Found via Robert Scoble’s Scobleizer News Pulse · a post on X, not independently verified Published 2026-11-19T17:30:00+00:00 Detected 2026-11-20T01:01:18+00:00
View original source

According to a post on X by @askalphaxiv, nearly 1,000 AI agents from 35 labs, including Stanford, MIT and ETH Zurich, ran research autonomously on alphaXiv's preprint server for three months. The agents produced 936 papers, each with code, data and weights. alphaXiv says the trial found significant errors in published human research and that it set up its own journal to review the output. The source snippet is cut off, so the full details of that journal and its review process are not available here.

AI-assisted summary based on the listed source.

For three months, nearly 1,000 agents from 35 labs, including Stanford, MIT and ETH Zurich, ran research autonomously on the server and produced 936 papers with code, data and weights; alphaXiv says they found significant errors in published human research and set up their own journal to review each o...

Researchers, preprint server operators and builders of AI research agents may want to watch how alphaXiv handles review of agent-written papers and the errors it reports in human research. The trial is a large-scale test case, but the claims come from alphaXiv itself and have not been independently checked by VQV. Next, look for the full report, details of the journal's review process, and whether the error findings are published with specifics.

Signal Strength 95% Technical label WATCH Public Interest 37 Category RESEARCH Reader Depth TECHNICAL

Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.

Public Interest components
Recognizable Entity Score 0 Practical Impact Score 36 Novelty Interest Score 94 Consequence Score 18 Curiosity Score 32 Shareability Score 51

VQV surfaced this signal because it is recent, relevant to AI Agents, connected to Scobleizer News Pulse (X).